The saying “what’s old is new again” doesn’t always apply to innovation and technology. But data compression seems to be getting a new turn at enterprise data “cool” from recent applications and research.

Here’s a sample of the varied, emergent attention on data compression in the last week: IBM calls compression a “must” and a “cornerstone” in terms of business big data programs; researcher Matt Mahoney posts an open benchmark to encourage compression research in AI and natural language processing; and Apple, that arbiter of consumer-side coolness, buys a compression specialist as part of its mobile efforts.

Whether its new applications of compression or just an existing answer to new data challenges, the small but consistent buzz begs a worthwhile question: is data compression regaining its cool?

From my vantage point, slick new approaches to compression have found traction with another long-standing element of data storage and control. For years, discussion of compression on the mainframe was a non-starter. It was typically very CPU intensive and the compression you’d want on a regular basis would also jack up the demand for capacity on the mainframe. There was also the cost of the software running on the mainframe, which adds up when you multiply that capacity.

Between a mix of data demands and, more importantly, sophisticated compression, businesses are finding they can revisit the possibility of compression on their mainframes without the threat of weighty CPUs or costly software. There’s also a security-side play, as you can’t pull, find and replace regular pattern strings from encrypted data. So, your best shot at compression comes before you start with encryption.

Download Free Trial:
SecureZIP

File Compression + Security

I’ll be presenting a compression webinar in conjunction with Computacenter twice in the coming weeks and you can also find me taking the stage to talk compression and encryption at Guide Share in Lahnstein, Germany later this month. In the meantime, what new ways are you diving in to data compression? Is it gaining traction with your mainframe and other legacy systems? Let me know in the comments below or on our LinkedIn page.